LG4AV: Combining Language Models and Graph Neural Networks for Author Verification

Stubbemann, Maximilian; Stumme, Gerd

doi:10.1007/978-3-031-01333-1_25

Search citation statements

Order By: Relevance

Paper Sections

Select...

Related Work1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2024

Publication Types

Select...

Article1

Relationship

Self Cite0

Independent1

Authors

Journals

Cited by 1 publication

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study presented in [14] addresses the authorship verification problem using a combination of a language model and a graph neural network called LG4AV. The approach calculates a vector representation for each author based on their article embeddings, which is then treated as a classification token ([CLS-a]) in the BERT language model.…”

Section: Related Workmentioning

confidence: 99%

Enriching Language Models with Graph-Based Context Information to Better Understand Textual Data

Roethel,

Ganzha,

Wróblewska

2024

Electronics

View full text Add to dashboard Cite

A considerable number of texts encountered daily are somehow connected. For example, Wikipedia articles refer to other articles via hyperlinks, or scientific papers relate to others via citations or (co)authors; tweets relate via users that follow each other or reshare content. Hence, a graph-like structure can represent existing connections and be seen as capturing the “context” of the texts. The question thus arises of whether extracting and integrating such context information into a language model might help facilitate a better-automated understanding of the text. In this study, we experimentally demonstrate that incorporating graph-based contextualization into the BERT model enhances its performance on an example of a classification task. Specifically, in the Pubmed dataset, we observed a reduction in balanced mean error from 8.51% to 7.96%, while increasing the number of parameters just by 1.6%.

show abstract

Section: Related Workmentioning

confidence: 99%